impr: Clean up `vectorOps` by aleksanderkatan · Pull Request #2848 · software-mansion/TypeGPU

aleksanderkatan · 2026-08-13T13:43:34Z

No description provided.

github-actions · 2026-08-13T13:46:58Z

Resolution Time Benchmark

---
config:
  themeVariables:
    xyChart:
      plotColorPalette: "#E63946, #3B82F6, #059669"
---
xychart
  title "Random Branching (🔴 PR | 🔵 main | 🟢 release)"
  x-axis "max depth" [1, 2, 3, 4, 5, 6, 7, 8]
  y-axis "time (ms)"
  line [0.97, 1.88, 4.24, 7.00, 7.46, 12.57, 23.64, 23.44]
  line [0.99, 1.89, 4.61, 6.57, 7.27, 10.57, 23.44, 23.40]
  line [0.96, 1.94, 3.88, 6.03, 7.42, 12.95, 23.88, 25.90]

---
config:
  themeVariables:
    xyChart:
      plotColorPalette: "#E63946, #3B82F6, #059669"
---
xychart
  title "Linear Recursion (🔴 PR | 🔵 main | 🟢 release)"
  x-axis "max depth" [1, 2, 3, 4, 5, 6, 7, 8]
  y-axis "time (ms)"
  line [0.28, 0.49, 0.68, 0.77, 1.05, 1.11, 1.34, 1.52]
  line [0.29, 0.51, 0.70, 0.82, 1.08, 1.16, 1.37, 1.55]
  line [0.34, 0.54, 0.69, 0.81, 1.11, 1.19, 1.43, 1.54]

---
config:
  themeVariables:
    xyChart:
      plotColorPalette: "#E63946, #3B82F6, #059669"
---
xychart
  title "Full Tree (🔴 PR | 🔵 main | 🟢 release)"
  x-axis "max depth" [1, 2, 3, 4, 5, 6, 7, 8]
  y-axis "time (ms)"
  line [0.87, 2.24, 3.82, 7.45, 12.34, 26.53, 55.40, 111.60]
  line [0.86, 2.22, 4.00, 7.09, 13.09, 26.79, 53.88, 110.27]
  line [0.79, 2.20, 4.29, 7.34, 12.83, 26.92, 55.66, 111.11]

github-actions · 2026-08-13T15:52:45Z

pkg.pr.new

packages
Ready to be installed by your favorite package manager ⬇️

https://pkg.pr.new/software-mansion/TypeGPU/typegpu@cf3e123016ac958d0c7bcab1f634686fcb77d5f8

benchmark
view benchmark

commit
view commit

github-actions · 2026-08-13T15:55:00Z

Bundle size comparison (`import * as ...` in PR vs `import * as ...` in target):

🟢 Decreased (max -29.83%)	➖ Unchanged	🔴 Increased	❔ Unknown
115	209	0	0

`import * as ...` in PR vs `import * as ...` in target (did bundle size increase?):

Click to reveal the results table (115 entries).

Test	tsdown
STATIC_allImports.ts	296.17 kB ($${\color{green}-4.4\%}$$)
tgpu_resolveWithContext.ts	270.50 kB ($${\color{green}-4.8\%}$$)
tgpu_bindGroupLayout.ts	270.50 kB ($${\color{green}-4.8\%}$$)
tgpu_mutableAccessor.ts	270.50 kB ($${\color{green}-4.8\%}$$)
tgpu_initFromDevice.ts	270.50 kB ($${\color{green}-4.8\%}$$)
tgpu_vertexLayout.ts	270.50 kB ($${\color{green}-4.8\%}$$)
tgpu_workgroupVar.ts	270.50 kB ($${\color{green}-4.8\%}$$)
tgpu_fragmentFn.ts	270.50 kB ($${\color{green}-4.8\%}$$)
tgpu_privateVar.ts	270.50 kB ($${\color{green}-4.8\%}$$)
tgpu_computeFn.ts	270.50 kB ($${\color{green}-4.8\%}$$)
tgpu_accessor.ts	270.49 kB ($${\color{green}-4.8\%}$$)
tgpu_comptime.ts	270.49 kB ($${\color{green}-4.8\%}$$)
tgpu_vertexFn.ts	270.49 kB ($${\color{green}-4.8\%}$$)
tgpu_resolve.ts	270.49 kB ($${\color{green}-4.8\%}$$)
tgpu_unroll.ts	270.49 kB ($${\color{green}-4.8\%}$$)
tgpu_const.ts	270.49 kB ($${\color{green}-4.8\%}$$)
tgpu_init.ts	270.49 kB ($${\color{green}-4.8\%}$$)
tgpu_lazy.ts	270.49 kB ($${\color{green}-4.8\%}$$)
tgpu_slot.ts	270.49 kB ($${\color{green}-4.8\%}$$)
tgpu_fn.ts	270.49 kB ($${\color{green}-4.8\%}$$)
STATIC_tgpu.ts	270.49 kB ($${\color{green}-4.8\%}$$)
STATIC_std.ts	96.96 kB ($${\color{green}-12.2\%}$$)
STATIC_d.ts	77.36 kB ($${\color{green}-13.6\%}$$)
std_abs.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_acos.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_acosh.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_asin.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_asinh.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_atan.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_atan2.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_atanh.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_ceil.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_clamp.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_cos.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_cosh.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_countLeadingZeros.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_countOneBits.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_countTrailingZeros.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_cross.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_degrees.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_determinant.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_dot4I8Packed.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_exp.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_exp2.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_extractBits.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_faceForward.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_firstLeadingBit.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_firstTrailingBit.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_floor.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_fma.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_insertBits.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_inverseSqrt.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_ldexp.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_log.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_log2.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_max.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_min.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_normalize.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_pow.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_quantizeToF16.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_radians.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_reflect.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_refract.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_reverseBits.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_round.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_saturate.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_sign.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_sin.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_sinh.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_smoothstep.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_sqrt.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_step.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_tan.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_tanh.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_transpose.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_trunc.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_distance.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_dot4U8Packed.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_fract.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_frexp.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_mix.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_modf.ts	55.86 kB ($${\color{green}-19.4\%}$$)
std_dot.ts	55.85 kB ($${\color{green}-19.4\%}$$)
std_length.ts	55.85 kB ($${\color{green}-19.4\%}$$)
std_rotateY4.ts	38.98 kB ($${\color{green}-23.7\%}$$)
std_rotateZ4.ts	38.98 kB ($${\color{green}-23.7\%}$$)
std_rotateX4.ts	38.97 kB ($${\color{green}-23.7\%}$$)
std_scale4.ts	38.97 kB ($${\color{green}-23.7\%}$$)
std_translate4.ts	38.97 kB ($${\color{green}-23.7\%}$$)
std_add.ts	38.14 kB ($${\color{green}-24.1\%}$$)
std_bitShiftLeft.ts	38.14 kB ($${\color{green}-24.1\%}$$)
std_bitShiftRight.ts	38.14 kB ($${\color{green}-24.1\%}$$)
std_div.ts	38.14 kB ($${\color{green}-24.1\%}$$)
std_mod.ts	38.14 kB ($${\color{green}-24.1\%}$$)
std_mul.ts	38.14 kB ($${\color{green}-24.1\%}$$)
std_sub.ts	38.14 kB ($${\color{green}-24.1\%}$$)
std_neg.ts	38.14 kB ($${\color{green}-24.1\%}$$)
std_bitcast.ts	35.90 kB ($${\color{green}-25.9\%}$$)
std_bitcastU32toF32.ts	35.89 kB ($${\color{green}-25.9\%}$$)
std_bitcastU32toI32.ts	35.89 kB ($${\color{green}-25.9\%}$$)
std_bitcastF32toU32.ts	35.89 kB ($${\color{green}-25.9\%}$$)
std_ge.ts	37.84 kB ($${\color{green}-29.8\%}$$)
std_gt.ts	37.84 kB ($${\color{green}-29.8\%}$$)
std_isCloseTo.ts	37.84 kB ($${\color{green}-29.8\%}$$)
std_le.ts	37.84 kB ($${\color{green}-29.8\%}$$)
std_allEq.ts	37.84 kB ($${\color{green}-29.8\%}$$)
std_eq.ts	37.84 kB ($${\color{green}-29.8\%}$$)
std_lt.ts	37.84 kB ($${\color{green}-29.8\%}$$)
std_ne.ts	37.84 kB ($${\color{green}-29.8\%}$$)
std_not.ts	37.84 kB ($${\color{green}-29.8\%}$$)
std_select.ts	37.84 kB ($${\color{green}-29.8\%}$$)
std_and.ts	37.83 kB ($${\color{green}-29.8\%}$$)
std_or.ts	37.83 kB ($${\color{green}-29.8\%}$$)
std_all.ts	37.83 kB ($${\color{green}-29.8\%}$$)
std_any.ts	37.84 kB ($${\color{green}-29.8\%}$$)

`import { ... }` in PR vs `import * as ...` in PR (is the library tree-Shakeable?):

Test	tsdown
tgpu_init.ts	260.96 kB ($${\color{green}-3.5\%}$$)
tgpu_initFromDevice.ts	260.42 kB ($${\color{green}-3.7\%}$$)
tgpu_resolve.ts	161.30 kB ($${\color{green}-40.4\%}$$)
tgpu_resolveWithContext.ts	161.23 kB ($${\color{green}-40.4\%}$$)
tgpu_bindGroupLayout.ts	62.30 kB ($${\color{green}-77.0\%}$$)
tgpu_mutableAccessor.ts	57.02 kB ($${\color{green}-78.9\%}$$)
tgpu_accessor.ts	57.02 kB ($${\color{green}-78.9\%}$$)
tgpu_privateVar.ts	55.71 kB ($${\color{green}-79.4\%}$$)
tgpu_workgroupVar.ts	55.71 kB ($${\color{green}-79.4\%}$$)
tgpu_const.ts	55.13 kB ($${\color{green}-79.6\%}$$)
tgpu_lazy.ts	54.93 kB ($${\color{green}-79.7\%}$$)
tgpu_fragmentFn.ts	39.65 kB ($${\color{green}-85.3\%}$$)
tgpu_fn.ts	39.60 kB ($${\color{green}-85.4\%}$$)
tgpu_vertexFn.ts	39.47 kB ($${\color{green}-85.4\%}$$)
tgpu_computeFn.ts	39.17 kB ($${\color{green}-85.5\%}$$)
tgpu_vertexLayout.ts	28.30 kB ($${\color{green}-89.5\%}$$)
tgpu_comptime.ts	15.91 kB ($${\color{green}-94.1\%}$$)
tgpu_unroll.ts	1.75 kB ($${\color{green}-99.4\%}$$)
tgpu_slot.ts	1.70 kB ($${\color{green}-99.4\%}$$)

If you wish to run a comparison for other, slower bundlers, run the 'Tree-shake test' from the GitHub Actions menu.

aleksanderkatan · 2026-08-14T10:19:52Z

-  }
-  throw new Error('Div called with invalid arguments.');
+  verifyKind([lhs, rhs], numericKind);
+  return generalizeFn((a, b) => a / b, upCast([lhs, rhs]));


Previously integer vectors used a separate function for integer division. This works because vector constructors cast their elements.

aleksanderkatan · 2026-08-14T11:11:55Z

@pullfrog review

pullfrog · 2026-08-14T11:11:59Z

Run failed. View the logs →

^{｜ Rerun failed job ➔ ｜ View workflow run ｜ via Pullfrog ｜ Using DeepSeek Flash (free via Pullfrog for OSS) ｜ 𝕏}

Copilot

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.

pullfrog

Important

This cleanup introduces two CPU regressions in the refactored paths: the deprecated vector bitcasts now return the wrong type and corrupted values, and isCloseTo crashes on mixed scalar/vector operands where the codegen path returns false. Both are pinned inline with confirmation from running the CPU implementations directly.

Reviewed changes — large cleanup that removes most hand-written per-kind overloads in data/vectorOps.ts in favor of a new data/generalizeFn.ts helper (generalizeFn / generalizeBoolFn / upCast), which applies an n-ary scalar function component-wise over vectors/matrices and rebuilds the result via a kind→constructor map:

New generalizeFn.ts with the applyArgs engine, kind/constructor maps, upCast, and the overloaded generalizeFn/generalizeBoolFn exports.
std/numeric.ts cpu* helpers (abs, acos, atan2, clamp, exp2, fract, log, max, min, mix, pow, sign, smoothstep, sqrt, trig, …) rewritten through generalizeFn.
std/operators.ts add/mul/div/mod/neg rewritten through generalizeFn (div/mod via upCast).
std/boolean.ts eq/lt/or/isCloseTo/select rewritten via generalizeBoolFn/generalizeFn.
Deprecated bitcastU32toF32/bitcastU32toI32/bitcastF32toU32 rewritten via generalizeFn.
VectorOps reduced to the non-generalizable set (all, length, dot, cross, matrix multiply, bit shifts); clamp renamed to clampScalar in numberOps.ts; tests updated for div-by-zero and distance.

ℹ️ Behavior change worth confirming

std.div (and std.mod) on integer vectors previously returned the operand on division by zero (divInteger returned lhs when rhs === 0); the generalized version now throws a Finite-Math-Assumption error via the u32/i32 cast. The rewritten test (forbids division by 0) intentionally encodes this, so it appears deliberate, but it is a silent behavioral change to the public CPU path. Worth confirming nothing downstream relies on the old lenient result.

ℹ️ Nitpicks

data/numberOps.ts still exports divInteger, which is now referenced nowhere after this refactor — it should be removed with the other deleted helpers (bitcast* impls were removed in the same file).

^{｜ Fix all ➔ ｜ Fix 👍s ➔ ｜ View workflow run ｜ Using DeepSeek Flash (free via Pullfrog for OSS) ｜ 𝕏}

pullfrog

Important

The isCloseTo crash fix (upcasting via upCast) resolves the CPU crash from the prior review but leaves a CPU/GPU divergence on mixed scalar/vector operands that the new test now locks in: the CPU computes a real component-wise result while the GPU codegen still returns the literal false.

Reviewed changes — incremental delta since the prior pullfrog review (9b5e1eb → cf3e123):

Reverted the deprecated bitcastU32toF32 / bitcastU32toI32 / bitcastF32toU32 back to their original VectorOps.* per-kind CPU impls, fixing the wrong-type/corrupted-value regression.
Removed the now-unreferenced divInteger helper from data/numberOps.ts (also addresses the prior nitpick).
Fixed the isCloseTo mixed scalar/vector CPU crash by upcasting operands via upCast before generalizeBoolFn, and expanded tests/std/boolean/isCloseTo.test.ts with close/distant/precision/mixed cases.

⚠️ CPU/GPU divergence on mixed `isCloseTo`

The new works for mixed types test asserts CPU results for mixed scalar/vector operands that a compiled shader cannot reproduce — the GPU codegen returns false for every mixed case while the CPU now computes the real comparison. Details inline.

^{｜ Fix all ➔ ｜ Fix 👍s ➔ ｜ View workflow run ｜ Using DeepSeek Flash (free via Pullfrog for OSS) ｜ 𝕏}

pullfrog · 2026-08-14T16:10:41Z

-    expect(isCloseTo(vec2h(0, 0), vec2h(0, 9), 10)).toBe(true);
-    expect(isCloseTo(vec2h(0, 0), vec2h(0, 11), 10)).toBe(false);
+  it('works for mixed types', () => {
+    expect(isCloseTo(d.vec2f(0, 0) as d.v2f | number, 0)).toBe(true);


The works for mixed types block asserts CPU results for mixed scalar/vector operands — isCloseTo(d.vec2f(0, 0), 0) → true — but the GPU codegen in std/boolean.ts still returns the literal false for mixed operands (the else branch when exactly one side is a scalar). A compiled shader evaluating this therefore yields false where the CPU returns true: a silent CPU/GPU divergence that the crash-fix did not resolve.

Technical details

# Mixed scalar/vector isCloseTo diverges CPU vs GPU ## Affected sites - packages/typegpu/tests/std/boolean/isCloseTo.test.ts:59-62 — new test asserts CPU-side mixed results (e.g. `isCloseTo(d.vec2f(0,0), 0)` -> `true`) - packages/typegpu/src/std/boolean.ts:338-348 — codegen `else` branch returns literal `'false'` whenever exactly one operand is a scalar snippet ## Required outcome CPU and GPU must agree on mixed scalar/vector `isCloseTo`. ## Suggested approach Either (a) make the GPU codegen broadcast the scalar to a matching vec and emit the real comparison (WGSL mixed scalar/vector arithmetic is valid — the codegen's own comment cites the spec), or (b) scope mixed operands out and have the CPU path (and this test) return `false` to match the codegen. Option (a) matches the upcast semantics now implemented on the CPU. ## Open questions for the human Is mixed scalar/vector `isCloseTo` an intended, supported semantic (fix the codegen), or out of scope (the CPU should match the GPU's `false`)?

aleksanderkatan added 15 commits August 13, 2026 11:17

POC

7d95880

Add type verification

b3edb44

Update sqrt

20caf75

Update trigo functions

2b8a614

Remove remaining simple unary ops

369a70f

Support matrices, remove remaining unary calls from vectorOps

998783e

Add binary op

14e08f8

Add verifyEqualTypes

2eeda52

Allow for upcasts

e616432

Rewrite simple component-wise ops

4407a2a

Add boolean mode

6119404

Simplify binary vector ops

c7d4704

Remove normalize from vectorOps

a270fe8

Accept an array of arguments

ce3a45a

Extract upcast, change mode to enum

a228ea2

aleksanderkatan linked an issue Aug 13, 2026 that may be closed by this pull request

impr: Clean up VectorOps #1336

Open

aleksanderkatan added 6 commits August 13, 2026 16:05

Prepare binaryuniformInput for generalization

7fcdf91

Make generalization work with any arity (except for types)

96719fa

Remove unary generalization

b45101a

Cleanup

fd1fb88

Add overload for 3 values

5a99b4e

Rename to generalizeFn

4144e69

aleksanderkatan added 5 commits August 13, 2026 18:00

Remove mix from vectorOps

75374f9

Remove select from vectorOps

46e52f0

Remove bitcasts from vectorOps

8c54d6a

Remove remaining simple vector ops

8ef19c5

Clean up vector ops imports

d4d56c0

aleksanderkatan added 2 commits August 13, 2026 18:33

Clean up vectorOps2 types

40cf5de

Fix test

2857f03

aleksanderkatan changed the title ~~Impr/clean up vectorops~~ impr: Clean up vectorOps Aug 13, 2026

aleksanderkatan added 6 commits August 14, 2026 10:28

Cleanup

446e12d

Generalize verifyKind for multiple arguments

779007e

Refactor generalizeFn.ts

50966bc

Simplify types

3367a2e

Simplify types

88423ed

Simplify types even further

a2dd774

aleksanderkatan commented Aug 14, 2026

View reviewed changes

aleksanderkatan added 3 commits August 14, 2026 12:33

More cleanup

080788a

Remove type verification

c3a3a00

Update snapshot

a953c5a

Merge remote-tracking branch 'origin/main' into impr/clean-up-vectorops

9b5e1eb

aleksanderkatan marked this pull request as ready for review August 14, 2026 12:46

Copilot AI lite review requested due to automatic review settings August 14, 2026 12:46

Copilot AI reviewed Aug 14, 2026

aleksanderkatan requested review from cieplypolar, iwoplaza, reczkok and vende11s August 14, 2026 12:46

pullfrog Bot reviewed Aug 14, 2026

View reviewed changes

Comment thread packages/typegpu/src/std/bitcast.ts Outdated

Comment thread packages/typegpu/src/std/boolean.ts

aleksanderkatan added 3 commits August 14, 2026 17:51

Cleanup divInteger

6f63e19

Revert bitcasts

2ca23b7

Fix isCloseTo

cf3e123

pullfrog Bot reviewed Aug 14, 2026

View reviewed changes

Navigation Menu

Uh oh!

Uh oh!

impr: Clean up `vectorOps` - #2848

impr: Clean up `vectorOps`#2848
aleksanderkatan wants to merge 41 commits into
mainfrom
impr/clean-up-vectorops

aleksanderkatan commented Aug 13, 2026

Uh oh!

github-actions Bot commented Aug 13, 2026 •

edited

Loading

Uh oh!

github-actions Bot commented Aug 13, 2026 •

edited

Loading

Uh oh!

github-actions Bot commented Aug 13, 2026 •

edited

Loading

Uh oh!

aleksanderkatan Aug 14, 2026

Uh oh!

aleksanderkatan commented Aug 14, 2026

Uh oh!

pullfrog Bot commented Aug 14, 2026 •

edited

Loading

Uh oh!

Copilot AI left a comment

Uh oh!

pullfrog Bot left a comment

Uh oh!

Uh oh!

Uh oh!

pullfrog Bot left a comment

Uh oh!

pullfrog Bot Aug 14, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Uh oh!

Uh oh!

Conversation

aleksanderkatan commented Aug 13, 2026

Uh oh!

github-actions Bot commented Aug 13, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Resolution Time Benchmark

Uh oh!

github-actions Bot commented Aug 13, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

github-actions Bot commented Aug 13, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Bundle size comparison (import * as ... in PR vs import * as ... in target):

import * as ... in PR vs import * as ... in target (did bundle size increase?):

import { ... } in PR vs import * as ... in PR (is the library tree-Shakeable?):

Uh oh!

aleksanderkatan Aug 14, 2026

Choose a reason for hiding this comment

Uh oh!

aleksanderkatan commented Aug 14, 2026

Uh oh!

pullfrog Bot commented Aug 14, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Uh oh!

pullfrog Bot left a comment

Choose a reason for hiding this comment

ℹ️ Behavior change worth confirming

ℹ️ Nitpicks

Uh oh!

Uh oh!

Uh oh!

pullfrog Bot left a comment

Choose a reason for hiding this comment

⚠️ CPU/GPU divergence on mixed isCloseTo

Uh oh!

pullfrog Bot Aug 14, 2026

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

github-actions Bot commented Aug 13, 2026 •

edited

Loading

github-actions Bot commented Aug 13, 2026 •

edited

Loading

github-actions Bot commented Aug 13, 2026 •

edited

Loading

Bundle size comparison (`import * as ...` in PR vs `import * as ...` in target):

`import * as ...` in PR vs `import * as ...` in target (did bundle size increase?):

`import { ... }` in PR vs `import * as ...` in PR (is the library tree-Shakeable?):

pullfrog Bot commented Aug 14, 2026 •

edited

Loading

⚠️ CPU/GPU divergence on mixed `isCloseTo`